{"type":"video","version":"1.0","html":"<iframe src=\"https://www.loom.com/embed/16688c0c3145466bae756293041ff443\" frameborder=\"0\" width=\"1920\" height=\"1440\" webkitallowfullscreen mozallowfullscreen allowfullscreen></iframe>","height":1440,"width":1920,"provider_name":"Loom","provider_url":"https://www.loom.com","thumbnail_height":1440,"thumbnail_width":1920,"thumbnail_url":"https://cdn.loom.com/sessions/thumbnails/16688c0c3145466bae756293041ff443-full-1695911607701.jpg","duration":448.03844650200017,"title":"Fine-tuning &lt;&gt; Inference Pipeline on Akash Network","description":"Hi, I'm Tom Montfort from Agora Labs, and in this video I will walk you through an example deployment on the Akash console. My co-founders and I have built a fine-tuning to inference pipeline on top of Akash, which allows ML engineers to take Open Source foundational models like LLaMa-2 and fine-tune them on their own proprietary data. We will show you how to select the model type, choose the dataset, and pass the necessary credentials. Additionally, we will demonstrate the process of creating deployments and launching them on an Akash provider. To close out, we discuss our future vision for this Akash project, including automatic redeployments and support for different model types."}